1. Field of the Invention
The present invention relates to the technology of checking used in converting product shape data generated by three-dimensional CAD, etc.
2. Description of the Related Art
In product development, etc., various analyses such as a thermal fluid analysis, etc. are conducted to evaluate a structure generated by three-dimensional CAD, etc. To perform these analyses, the product shape data generated by three-dimensional CAD, etc. are converted into an analytic model appropriate for the analyses, and the analyses are conducted on the analytic model.
FIGS. 1A, 1B, 1C and 1D show examples of converting a three-dimensional model based on the product shape data into an analytic model.
These figures show examples of converting a structure of a three-dimensional model based on the product shape data generated by three-dimensional CAD into an analytic model. FIG. 1A shows the front view of a three-dimensional model before conversion. FIG. 1B shows the reverse side of the three-dimensional model before conversion. FIG. 1C shows the front view of the three-dimensional model after conversion. FIG. 1D shows the reverse side of the three-dimensional model after conversion.
An analytic model is generated by dividing an original three-dimensional model by assigning a grid pattern to it and determining the validity of each rectangle element. Therefore, as shown in FIG. 1, an analytic model has an error between the original three-dimensional model in shape, area, volume, etc.
Since an analytic model obtained by conversion has an error in shape, area, volume, etc., it is probable the analytic model cannot be used in an analyzing process depending on the level of the error. The determination as to whether or not the analytic model can be used in the analyzing process has been conventionally made based on the empirical rule of an operator.
That is, conventionally, three-dimensional models before and after conversion have been displayed as shown in FIG. 1, and an operator visually checks the state of its shape, and sensuously determines the acceptability of the analytic model. Therefore, for lack of quantitative determination standards, and due to personal discretion, the subsequent analysis result largely depends on the operator who normally continues the process. As a result, there has been the demerit that the accuracy (reliability) of the analysis result is affected by the sense of each operator.